Structuring Category Analysis Using Statistical Technique

Abstract

The transportation-planning process incorporates the development of representative-trip-generation models from base year data for making forecasts of travel in subsequent years. Category analysis is a technique based on the unit of the household and its characteristics. The data are cross-classified over various socioeconomic schemes. Trip rates based on such schemes are made for a particular category of analysis unit. When the household characteristics are cross-classified with the available data, certain cross-classification cells may get sparse data and certain others may not get any data. Under such circumstances, conventional application of category analysis may not be appropriate. To overcome this difficulty a linear statistical model framework was developed. The estimated mean trip rates for the cells of the cross-classified table use a model fit based on data from all cells and not just the data from any given cell, which may not be reliable. Developed models were illustrated for the different cross-classification schemes.